eRata
`
`ea
`
`Apple Inc. EX1022 Page 1
`
`SND ETN
`
`IPR2018-01474
`
`IPR2018-01474
`Apple Inc. EX1022 Page 1
`
`

`

`McGnAW-rILL ENCYCLOPEDIA OF
`
`NINECHING
`
`SECOND EDITION
`
`Sybil P. Parker
`Editor in Chief
`
`McGraw-Hill, Inc.
`Bogota
`Auckland
`San Francisco Washington, D.C.
`New York
`Caracas
`Lisbon
`London Madrid Mexico City Milan Montreal
`New Delhi
`San Juan
`Singapore
`Sydney
`Tokyo
`Toronto
`
`IPR2018-01474
`Apple Inc. EX1022 Page 2
`
`IPR2018-01474
`Apple Inc. EX1022 Page 2
`
`

`

`Most ofthe material in this volume has been published previously in the
`McGRAW-HILL ENCYCLOPEDIA OF SCIENCE & TECHNOLOGY,
`Seventh Edition, copyright © 1992 by McGraw-Hill, Inc., and in the
`McGRAW-HILL YEARBOOK OF SCIENCE & TECHNOLOGY, copyright
`© 1992, 1991, by McGraw-Hill, Inc. All rights reserved.
`
`McGRAW-HILL ENCYCLOPEDIA OF ENGINEERING, Second Edition.
`Copyright © 1993 by McGraw-Hill, Inc. All rights reserved. Printed in the
`United States of America. Except as permitted under the United States
`Copyright Act of 1976, no part of this publication may be reproduced or
`distributed in any form or by any means, or stored in a database orretrieval
`system, without the prior written permission of the publisher.
`
`Pont VOT One
`
`DOW/DOW
`
`987654
`
`Library of Congress Cataloging in Publication data
`
`I. Parker, Sybil P.
`
`I]. McGraw-Hill
`
`McGraw-Hill encyclopedia of engineering / Sybil P, Parker, editor in
`chief. — 2nd ed.
`cm.
`Includes bibliographical references and index.
`ISBN 0-07-051392-9
`1, Engineering—Encyclopedias.
`Inc.
`TA9.M36
`620’ .003—dce20
`
`1993
`
`92-43106
`CIP
`
`ISBN 0-07-051392-9
`
`IPR2018-01474
`Apple Inc. EX1022 Page 3
`
`IPR2018-01474
`Apple Inc. EX1022 Page 3
`
`

`

`Digital computer
`
`output
`
` Bourdon G-tube
`core
`
`
`mounting block
`
`
`
`
`
`
`see
`Fig.
`2. Bourdon-tube pressure transducer employing a linear
`variable differential transformer (LVDT). (After E. a =
`Herceg, Handbook ofMeasurement and Control, Schae
`Engineering, 1976)
`
`is zero. This is the balance of null position,
`output
`Whenthe core is displaced from the null point,
`the
`two secondary voltages are no longer alike and the
`transformer produces an output voltage. With proper
`design,
`the output voltage varies linearly with core
`position over a smal! range. Motion of the core in the
`opposite direction produces a similar effect with 180
`phase reversal of the alternaling output voltage.
`trans-
`The principal advantages of the differential
`former over other displacement transducers, such as
`the resistance potentiometer, are absence of contacts
`and infinite resolution. No friction is introduced by
`the measurement, and movement smaller than a mi-
`croinch (25 nanometers) can be sensed. The separa-
`tion between coil and core makes the differential
`transformer useful in difficult and dangerous environ-
`ments. Stability of the null makes it ideal as a null
`sensor
`in self-balancing devices and servomecha-
`nisms. Typical applications are machine tool Inspec-
`tion and gaging, pressure measurement (Fig. 2), lig-
`uid level control,
`load cells, and Eyroscopic instru-
`ments.
`_The linear variable differential transformer (LVDT;
`Fig. 3) is the commercially prevalent form. A rota
`anolderdevice.Isincnty eaFig. 4
`abe
`ee
`ily 18 NOL as good, and its
`principal use is as a null sensor, Both translati
`and rotational E-pickoffs are
`m;
`eel
`i
`made.
`
`Primarycoil
`_
`
`;
`ees 2, secondary
`
`coil 1, secondary
`
`motion to be
`indicated or4
`controlled *
`
`> to ac-valtape
`
`Source (constant)
`
`“orm=
`¥
`
`difference volt:age
`,
`OULPUt Exec. 4 =r sec_2
`Fig. 3. Linear variable differential tansforme avpr
`rr
`7m
`
`The amplitude of the ac output Voltape forms .
`eV
`shaped curve when plotted against core
`q
`POSitign: the
`phase angle abruptly reverses by | gq
`a the ny
`oint. When the bottom of the V-curye is ¢
`if
`in closer detail,
`it
`is seen that
`the outpuy Vol Ini
`balance is not exactly zero. The small Tesidual
`voltage consists of higher harmonies Ofthe inp ul
`quency, as well as a fundamental frequency ie te
`nent 90 degrees out of phase (called the qUadrajy,
`component).
`—
`fe
`Electronic signal conditioning is comm
`only {m.
`ployed to eliminate the residual and (o
`Make thy
`
`
`
`
`ay
`Utput from
`Secondary
`0
`windings
`
`
`
`
`constant ac
`voltage input
`Fig. 4. E-shaped differential transformer.(After P. J.
`O'Higgins, Basic Instrumentation, McGraw-Hill, 1966)
`
`transducer usable with standard de instrumentation
`The electronic circuit consists of an ae oscillator (cu
`rier generator) to drive the input windings, plus ade-
`modulator and an amplifier to convert the output into
`de. The microelectronics can be built
`into the trans
`former housing. and the resulting package is sold &
`ade-LVDT. Sc TransrorMer.
`‘
`Gerald Weis
`Bibliography. W. R. Ahrendt and C, J. Savant, Jt,
`Servomechanism Practice. 1960; E. Q. Doebelit
`Measurement Systems, 3d ed.
`|982; E. E. Hettet
`Handhook of Measurement and Control, 1976,
`
`——————
`Digital computer
`jon’
`Any device for performing mathematical calcula
`On numbers represented digitally, by extensio", .
`device for manipulating symbols according 10 @ a
`tailed procedure or recipe. The class of digital i
`puters includes microcomputers, conventional pre
`machines and calculators, digital controllers
`s,
`Industrial processing and manufacturing ae
`Store-and-forward data communication equim
`and electronic data-processing systems.
`(j-pro-
`In this article emphasis is on electronic str it
`Bram digital computers, These machines Str 3 0
`nally many thousands of numbers or otherH&M st
`Information, and control and execute complicalé ip
`quences of numerical calculations and other ee
`lations on this information in accordance with ign
`tions also stored in the machine. The first ee
`this article discusses digital system fundament® at
`viewing the components and building olori
`Which digital systems are constructed. The fol it
`Section introduces the stored-program generalPret
`computer in more detail and indicates the ls
`
`IPR2018-01474
`Apple Inc. EX1022 Page 4
`
`IPR2018-01474
`Apple Inc. EX1022 Page 4
`
`

`

`Digital computer
`
`285
`
`Table 2. American standard alphabetic codefor binary
`representation ofletters
`
`11000001
`11000010
`01000011
`11000100
`01000101
`01000110
`41000111
`41001000
`01001001
`01001010
`41001011
`01001100
`41001101
`
`A
`B
`Cc
`D
`E
`F
`G
`H
`|
`J
`K
`L
`M
`
`
`
`14001110
`01001111
`11010000
`01010001
`01010010
`174010011
`01010100
`41010101
`11010110
`01010114
`01011000
`44011001
`41011010
`
`N
`oO
`P
`Q
`R
`S$
`T
`U
`Vv
`Ww
`«xX
`Y
`Z
`
`final section traces the history of stored-program dig-
`ital computer systems and shows howthe require-
`ments of new applications and the development of
`new technologies have influenced system design.
`DicitaL System FUNDAMENTALS
`A digital system can be considered from many
`points of view. At the lowest level it
`is a network of
`wires and mechanical parts whose voltages and posi-
`tions convey coded information. At another level it is
`a collection of logical elements, each of which em-
`bodies certain rules, but which in combination can
`carry out very complex functions. At a still higher
`level, a digital systemis an arrangementof functional
`units or building blocks which read (input). write
`(output), store, and manipulate information.
`Codes. Numbers are represented within a digital
`computer by meansofcircuits that distinguish various
`discrete electrical signals on wires inside the machine.
`Theoretically, a signal on a wire could be made to
`represent any one ofseveral different digits by means
`of the magnitude ofthe signal. (For example, a signal
`from 0 to | V could represent the digit zero, a signal
`between | and 2 V could represent the digit one, and
`so on up to a signal between 9 and 10 V.
`the digit
`nine.) In practice,
`the most reliable and economical
`circuit elements distinguish between only two signal
`levels, so that a signal between 0 and 5 V may rep-
`resent
`the digit zero and a signal between 5 and 10
`V,
`the digit one. These two-valued signals make it
`necessary to represent numbers and symbols using a
`corresponding base-two or binary system. Table 1
`lists the first 20 binary numbers and their decimal
`equivalents,
`Data are stored and manipulated within a digital
`computer in units called words. The binary digits
`(called bits), which make up a word, mayrepresent
`either a binary number or a collection of binary-coded
`alphanumeric characters. For example, the two-letter
`word *‘it’’ may be stored in a 16-bit computer word
`as follows, making use of the code shown in Table 2:
`0100100101010100
`
`The computer word merely contains a binary pattern
`
`of alternating |'s and 0’s, and it is up to the computer
`user to determine whether that word should be inter-
`preted as the English word “‘it'’ or as the decimal
`number 18,772.
`Logical circuit elements. Two kinds oflogical cir-
`cuits are used in the design and construction ofdigital
`computers: decision elements and memory elements.
`A typical decision element provides a binary output
`as a function of two or more binary inputs. The AND
`circuit,
`for example, has two inputs and an output
`which is | only when both inputs are |. A memory
`element stores a single bit of information and is set to
`the | state or reset to the 0 state, depending on the
`signals on its input lines. And because such a circuit
`can be caused alternately to store O's and 1's from
`lime to ume, a memory element is commonly called
`a flip-flop.
`that are
`These two basic logical elements are all
`required to construct the most elaborate and complex
`digital arithmetic and control circuits. A simple ex-
`ample of such a circuit is shown in Fig. 1. Here the
`object is to perform a simple binary count, as shown
`in the table at the bottom of Fig.
`|. As long as control
`signal C is equal ta 1,
`the counting continues. When
`the control
`input
`is 0,
`the counter is to remain in
`whatever state it had last counted to. Two flip-flops
`are used,
`labeled QI and Q2, and will be made to
`count
`through the sequence 0,1,2,3,0,1, .. .
`. To
`understand the design, it is necessary to introduce one
`more concept,
`the complementary output of a flip-
`Table1.Counting from 0 to 19bydecimal and
`flop. Each flip-flop generally has two outpul wires,
`
`binarynumbers. =
`which are always of opposite polarity, When flip-flop
`Q1 is storing a 1, output Q1 is| and the complemen-
`Decimal number
`Binary number
`lary output (which is labeled @1 and pronounced Q1
`bar) is 0. When the flip-flop contains a 0,
`the Q1
`00
`output is | and the Q1 output is 0.
`a1
`To analyze the circuit, note first that, when control
`o2
`03,
`input C is 0, the outputs of all AND gates are 0 and,
`04
`because the reset and set inputs to both flip-flops are
`05
`0, the flip-flops will remain in whateverstate they last
`06
`reached. Now suppose that Q1 and Q2 both contain 0
`O7
`08
`and that the control
`input becomes |. While flip-flop
`Q2 contains a 0,
`its Q2 output is also 0 and AND gale
`number | (labeled AND 1) is effectively turned off so
`that
`the reset and set inputs to Q1 are both 0. Thus
`flip-flop @1 will remain in the 0 state. For the same
`reason AND gate 4 will also be turned off, and the
`reset input to flip-flop Q2 will be 0.However, from
`flip-flop @2 complementary output Q2 will be in the
`j state, and (while the control input is 1) AND gate 5
`will be turned on and the set input to Q2 will be I.
`Flip-flop @2 will thus be turned on by the first clock
`
`00110
`
`10Bi
`
`IPR2018-01474
`Apple Inc. EX1022 Page 5
`
`
`
`‘ a fF &i& iT
`
`SSe
`
`IPR2018-01474
`Apple Inc. EX1022 Page 5
`
`

`

`286
`
`Digital computer
`
`control (C)
`
`Binary count:
`Ql
`Q2
`0
`0
`0
`1
`as
`1
`1
`
`ay
`0
`
`8
`1
`
`Logic equations:
`SOl =C-Q2+01
`
`RQ1 =C- Q2- Ql
`soz =: 02
`
`RQ2 = C+Q2
`
`Fig, 1, Simple digital counting circuit.
`
`Note; C- Q2 means CAND Q2
`
`pulse to occur after C is turned on; and from one
`clock pulse time to the next
`the two flip-flops will
`change from the (0,0) state to the (0,1) state. A care-
`ful review of the indicated circuits will show that the
`counter will indeed go through the count sequence as
`shown, as long as the control
`input is 1. The logic
`equations in Fig.
`| represent another way of describ-
`ing the circuit and may be used in place of the more
`cumbersome diagram.
`Physical components. The logical elements de-
`scribed in the paragraphs above are the fundamental
`conceptual components used in virtually all digital
`systems. The actual physical components which were
`used to realize conceptual gates and flip-flops in some
`specific piece of equipment are dependent on the sta-
`tus of electronic technology at the time the equipment
`was designed.
`In the 1950s the earliest commercial
`computers used vacuum tubes, resistors, and capaci-
`tors as components. A flip-flop typically required a
`dozen or more such components in these
`first-gener-
`ation computers. Betweenthe late 195()s
`and middle
`1960s, solid-state transistors and diodes replaced th
`:
`e
`vacuum tubes, and the resulting second-generatio
`systems were considerably more reliable than their
`first-generation predecessors:
`(h
`€y were also smaller
`and consumed less
`power. But the number of elec-
`tronic components
`per conceptual logical component
`remained aboutthe
`Same—a dozen or more for a flip-
`flop.
`Since the mid-1960s the integrated circuit (IC) has
`ding block for digital
`
`nents permitted designers of the
`early third-generation
`systems to provide much more
`ili
`1
`1
`Capability
`per
`nent than was possible with the first- ee ie race
`eration technology. Since the
`citeeey
`aoe
`mid-1960s
`inteor:
`circuit
`technology has consistently ae> an
`
`large-scale-integration (LSH) citreus
`typical
`5 Contain
`thousands offlip-flops and gates,
`System building blocks. On a completely differen,
`conceptual level, a digital compuler can be regarde
`as being composed of functional,
`system buildin
`blocks, containing (among other things) subassen”
`blies of the fundamental logical components, A com.
`puter viewed at
`this level may be described jn i
`oversimplified fashion by the diagramofFig, 2. The
`computation and control block (often called the cen.
`tral processing unit, or CPU) IS constructed entirely
`of logical elements of the kind described above. The
`main memory, Which may store froma few thousand
`to several million binary digits, andthe inpuvoupy,
`and auxiliary memory devices (the so-called periph.
`eral equipment) are specialized devices that aré avail
`able over a range of speeds and operating character.
`istics.
`Main memoryis a building block capable ofstoring
`data or instructions in bulk for use by the computation
`and control portion of the computer, The importam
`characteristics of a memoryare capacity, access time,
`and cost. Capacity 1s the amountofdatathat the com.
`puter can store. Access time is the maximum interval
`between a request
`to the memoryfor data and the
`moment when the memory can provide that data, Cost
`is measured by dividing total memory cost bythe
`number of bits stored. For first-generation systems,
`designers used a variety of technologies in realizing
`main memory: mercury delay lines, electrostatic stor-
`age tubes, and magnetic drumsall appeared in vanous
`products. But second- and third-generation systems
`were almost exclusively built using magnetic core
`main memories, Starting in the early 1970s,
`the inte-
`grated circuit memory was introduced, and 1s nowthe
`most widely used technology.
`Input/output and auxiliary memory peripherals rep-
`resent
`the other major computer building blocks.
`Equipment is now and has from the beginning been
`available for
`feeding information to the computer
`from paper tape and punched cards, and for recelvin:
`data from the computer andprinting it, or punching
`on tape cards, But in the intervening years, designtts
`have provided additional output devices which record
`computer data on microfilm, or plot data on graphs
`or use data to control physical devices such as valves
`or rheostats. They have also designed input ge
`
`computation
`and
`control
`
`(peripherals)
`
`input
`
`output
`
`auxiliary
`memory
`
`Fig. 2. Block diagram of a digital computer.
`
`IPR2018-01474
`Apple Inc. EX1022 Page 6
`
`IPR2018-01474
`Apple Inc. EX1022 Page 6
`
`

`

`Digital computer
`
`287
`
`5 bits
`11 bits[
`
`
` command address
`Fig, 3. Sixteen-bit instruction.
`
`ment which feeds the computer data fromlaboratory
`instruments, and from devices which scan documents
`and ‘‘read’’ printed characters, Data can be transmit-
`ted to and from the computer over ordinary telephone
`lines, and a wide variety of devices generally called
`terminals make it possible for people to send data to,
`or receive requested data from, a computer system
`located hundredsor thousands of miles away.
`The earliest auxiliary memory equipment recorded
`data on reels of magnetic tape. Magnetic tape units
`are still very widely used, for although they are slow
`in comparison to the operating speeds of modern
`computers—it typically takes 2-30 min to read all the
`data on a 2400-ft (732-m) reel of tape, depending on
`the speed of the tape untt—they makeit possible to
`store large volumes of data at low cost by virtue of
`the low cost ofthe tape itself. The other widely used
`auxiliary memory devices are the magnetic disk and
`drum, both of which provide faster access to data than
`do the tape units, but at higher cost per bit of data
`stored.
`
`Storep Program Computer
`
`implements the instructions, Because the instructions,
`like the dala, are stored in computer words, one be-
`gins by examining how an instruction 1s stored in a
`word, As an example, assume one ts looking al a
`small computer with words 16 bits long, and assume
`further that an instruction is organized as shown in
`Fig. 3. In this simple computer an instruction has two
`parts: the first 5 bits of the word specify which of the
`computer's repertoire of commands is to be carried
`out, and the last || bits generally specify the address
`of the word referred to by the command. A 5-bit
`command permits up to 32 different kinds of instruc-
`lions in the computer, and an 11-bit address permits
`one to address up to 2048 different memory locations
`directly.
`Components and building blocks described in the
`instruction types for a computer of this
`Typical
`kind are listed below.
`preceding paragraphs could be organized in a multi-
`tude of different ways. The first practical electronic
`Load. Load the number from the prescribed mem-
`ory location into the arithmetic unit.
`computers, constructed during the latter pan’ of World
`Store. Store the number fromthe arithmetic unit in
`War I], were designed with the specific purpose of
`computing special mathematical functions. They did
`the memory at the prescribed memory location.
`their jobs very well, but even while they were under
`Add. Add the contents of the addressed memory lo-
`cation to the number in the anthmetic unit,
`leaving
`construction, engineers and scientists had come to re-
`the result in the arithmetic unit
`alize that it was possible to organize a digital com-
`the addressed
`Subtract. Subtract
`the contents of
`puter in such a way that
`it was not oriented toward
`memory location from the number in the arithmetic
`some particular computation, and could in fact carry
`unit, leaving the result in the arithmetic unit,
`out any calculation desired and defined by the user
`Branch.
`\{ the numberin the arithmetic unit is zero
`The basic machine organization invented and con-
`or positive, read the next instruction from the address
`structed af
`that
`time was the stored-program com-
`puter, and it continues to be the fundamentalbasis for
`in the next-instruction register as usual. If the number
`each of the hundreds of thousands of computing sys-
`in the arithmetic unit
`is negative, store the address
`from the branch instruction itself in the next-instruec-
`temsin use today.
`It has also become a system com-
`ion register, so that the next
`instruction cared oul
`ponent, since the microcomputer Is simply a stored
`will be retrieved from the address givenin the branch
`program computer onasingle integrated-circu)t chip.
`instruction.
`The concept of
`the stored-program computer
`is
`Halt. Stop; carry out no further instructions until
`simple and can be described with reference to Fig. 2.
`Main memory contains,
`in addition to data and the
`the operator presses the RUN switch on the console,
`Input. Read the next character from the paper tape
`results of intermediate computations, a set of instruc-
`reader into the addressed memory location and then
`tions (or orders, or commands,as they are sometimes
`move the tape so a new characteris ready to be read.
`called); these specify how the computeris to operate
`Output. Type out the character whose codeis stored
`in solying some particular problem. The computation
`in the right-hand half of the addressed memory loca-
`and contro] section reads these instructions from the
`tion.
`memory one by one and performs the indicated oper-
`alions on the specified data. The instructions can
`control
`the reading of data from input or auxiliary
`memory peripherals, and (when the prescribed com-
`pulations are completed) can send the result to auxil-
`lary memory, or to output devices where it may be
`Printed, punched, displayed, plotted, and so forth.
`The feature that gives this form of computer organi-
`Zallon its great power is the ease with which instruc-
`ons can be changed; the particular calculations car-
`nied out by the computer are determined entirely by a
`Sequence of
`instructions stored in the computer's
`memory: that sequence can be altered completely by
`Simply reading a new set of instructions into the
`memory through the computer input equipment.
`Instructions. To understand better the nature ofthe
`Stored-program computer, consider in more detail the
`Of instructions it can carry out and the logic of
`the computation and control unit which interprets and
`
`the
`With the exception of the branch command,
`preceding instructions are easy to interpret and to un-
`derstand. The load and store commands move data to
`and from the arithmetic unit, respectively. The add
`and subtract commands perform arithmetic opera-
`tions, each using the number previously left
`in the
`arithmetic unit as one operand, and a number read
`from a designated memory location as the other. The
`halt command simply tells the computer to stop and
`requires intervention by the operator to make the
`compuler initiate computation again, The input and
`output commands make possible the reading of infor-
`mation into the computer memory from a paper tape
`in? device. and the printing out of the results from
`previous computations on an output typewriter.
`To understand the branch command, consider how
`the computation and control unit of Fig. 4 uses the
`instructions in the memory. To begin with,
`the in-
`
`IPR2018-01474
`Apple Inc. EX1022 Page 7
`
`IPR2018-01474
`Apple Inc. EX1022 Page 7
`
`

`

`288
`
`Digital computer
`
`|
`
`control logic
`
`|
`
`mee
`
`from memory
`
`Fig. 4, Computation and contro! unit.
`
`operand
`
`from/to memory
`
`address to
`
`contro! memory
`
`logic causes the addre
`however, the control
`SS in the
`ney,
`instruction register to be transferred to the
`struction address register before going on to the tole
`step. The computer will
`then CONTINUE With one g.
`quence of commandsif the previous arithmetic result
`was positive. and with anotherj| the result was ma
`ative, This seemingly simple operation is one of the
`next-instruction
`address registe
`most
`important features of a Computer.
`It gives ih.
`wokres:
`instruction register
`computer a decision-making capability that permits
`
`command address:5aE
`to examine some data, computea result, and COmtinie
`SS SS
`with one of two sequences of calculations o, Opera.
`lions, depending only on
`the Computed resy;
`Resume. As the fourth and final step in the bas
`quence, When the command has beeninterpreted and
`carried oul properly, the control
`logic returnsto the
`read step and repeats the entire series ofsteps,
`A sequence of instructions intended to cary oy
`some desired function Is called a program;collections
`of such programsare called software (as distinguished
`from the equipment, or hardware), and the actof pre.
`paring such programsis called programming, Because
`a computer can perform no useful
`function unyj
`someone has written a program embodyingthat func-
`tion, the programming activity is an exceedingly im-
`portant one and provides a basic limitation tothe fa-
`cility with which the computer can be applied to new
`areas.
`
`structions which are to be carried out must be stored
`in consecutive storage locations in memory. Assume
`that the first of a sequence of commandsis in memory
`location 100, Then the "‘next-instruction address reg-
`ister’’ in the computation and contro! unit (Fig. 4)
`contains the number 100, and the following sequence
`of four events takes place: (1) read, (2) readdress, (3)
`execute, and (4) resume.
`instruction
`Read. The control
`logic reads the next
`to be carried out
`from the memory location whose
`address is given by the next-instruction address reg-
`ister. The instruction coming from memory is stored
`in another register called the instruction register.
`(In
`this example the next-instruction address
`register
`started out containing the number 100, and so the in-
`struction in memory location 100 is transferred to the
`instruction register. }
`Readdress. The control logic nowadds unity to the
`numberin the next instruction address register. (In the
`present example this changes the number in the next
`instruction address register from 100 to 101. The re-
`sult is that, when the computer has interpreted and
`carmied out the instruction from location 100, follow-
`ing the rules given in the third and fourth steps below
`it will return to the read step above and next interpret
`and carry out the instruction from location 101.)
`| Execute. The instruction from location 100 is now
`Pee wena and must be carried out.
`gic
`first looks at the command portion
`of the instruction in the lefi-hand 5 bits of the register
`and interprets or decodesit to determine what to d
`next. If the instruction is add, subtract,
`load, or a
`put, the control logic first uses the address in the i
`:
`struction register—the
`address
`i
`ities
`8
`address associated w h
`th
`command—and
`reads
`th
`Tae
`ipbdean
`a
`reads the word from that addressed
`location in Memory;it then proceeds to load the word
`into the arithmetic unit, add it to or subtract it
`f
`the number inthe arithmetic unit. or transfa Pian
`output typewriter, depending on the co:
`
`Computer characteristics. A computer installation \s
`complex. Consequently it
`is difficult
`to describe a
`system or to compare the characteristics of lwosys-
`tems without
`listing their instruction types and de-
`scribing their modes of operation at some length
`Nevertheless, certain important descriptors are com-
`monly used for comparison purposes and are shown
`in Table 3. where salient characteristics of wo typ
`cal systems are shown. Definitions of these character-
`istics can be stated as follows.
`Memory cycle time is the time required to read 3
`word from main memory. Most modern compuless
`have integrated-circuit memories with cycle times It
`the ranges shown in Table 3. Add time is the time
`required to perform an addition,
`including the ume
`necessary to extract the addition imstruction itself and
`the operand from memory. Main memory storage
`pacity is the number of words of storage available
`the computation and control unit. Typically. 9c"
`puler manufacturer gives the buyer some loi
`buyer can purchase enough memory to meet the nee
`of the expected application. This internal capacity FF
`fers to the high-speed internal storage only, and me
`not include disks, drums, or magnetictape.
`i
`Word length is the number ofbits in a compl
`word.
`id
`System cost may vary over a range of 5 oF aes *
`‘0 |
`for a particular computer because of the gr
`
`:
`i
`»
`the c
`parol Wea
`Simply prevents all further Operation
`ee the operator console.
`the commandis branch, the controllogic begins
`aesfl the numberin the arithmetic unit.
`If that
`fO OF positive, the control logic ¢
`0 the fourth step below. If the number i Baia
`
`OeiFat
`pug a
`aaa time
`126,000
`ReanOy Storage
`Word length
`4:000,000 words: 6 is
`
`IPR2018-01474
`Apple Inc. EX1022 Page 8
`
`IPR2018-01474
`Apple Inc. EX1022 Page 8
`
`

`

`Digital computer
`
`289
`
`§'
`
`«
`‘
`
`bh
`
`
`data
`study
`|
`results
`jnputoutput
`"
`jpeg?
`:
`
`[/——_—_4
`input
`program
`
`K4 Fa E44 HY OH OR
`input
`output
`data
`data
`Sta oS
`
`co
`tion
`| compute
`
`@ ‘
`
`-———
`time
`operator
`|
`eperator
`tt
`preparation
`aaaae
`!
`preparation
`study
`results I
`"inputoutputbapa}oe—HIjo
`ipa
`input
`output
`input
`input
`output
`program
`data |
`data |
`program Il data I
`data I
`|———_________
`foo
`If
`0
`
`_ computation
`©
`
`‘operators PP
`i
`continuous activity by many independent operators
`
`program and
`data |
`
`program and
`data II
`
`intermediate
`results |
`
`program and
`data I
`
`
`
`output eeerefeel aetaniead
`
`store intermediate
`output
`output
`results J
`data I
`data |
`
`computation a eS a
`(0)
`:
`interrupt
`i
`I
`i
`
`Fig. 5. Comparison of efficiency for three generations of computers: (a) first, (b) second, and (c) third.
`
`variety of options offered the buyer by the manufac-
`turer—options such as memorysize, special
`instruc-
`tions for efficiency in certain calculations. and num-
`ber and type of peripheral devices.
`ae are obviously a number of other measures
`ei
`ay be used todescribe a computer. Theyin-
`such characteristics
`as multiplication time.
`ae ae between inpuVoutput equipment and
`avail a physical size, power consumption, and the
`“allability of a variety of computing options and spe-
`cial features.
`.
`
`Evo.ution oF CAPABILITIES
`j
`his process by which new circuil and peripheral
`aes technologies led to the development ofa
`ie generations of computers was discussed
`ogy th ut simultaneous with the changes in technol-
`tureofow Came changes in the structure or architec-
`improvehee These changes were introduced to
`Besignen Capability and efficiency of systems, as
`came
`;
`A
`fe
`actually used
`to understand how computers were
`coe efficiency. One way oflooking at sys-
`ation of 4 a is indicated in Fig. 5, where the oper-
`following fo mputer is shown broken downinto the
`ACLiVities as ‘a parts. (1) Operator time includes such
`Magnetic tg eee cards into a card reader, loading
`computer«. Onto a tape unit, setting up controls on
`Tesults, Reece S panel, and reviewing printed
`Tal devices Put comes to the computer from periph-
`ude ingt of from auxiliary memory. The inputs
`*"ructions from the operator, inputs of pro-
`
`grams to be run, and inputs of data. (3) Computation,
`being the principal activity, should occupyrelatively
`much ofthe total time. (4) Output includes storage of
`intermediate and final
`results in auxiliary memory,
`and printing of results along with instructions or
`warnings to the computer operator.
`First-generation computer.
`In the first generation
`of computer equipment only one of these activities
`could be carried out at a time. Between jobs the com-
`puter was idle while an operator made ready for the
`next task. When the operator was ready, the program
`was read into the computer from some input device
`and the input data were then loaded. The program
`operated upon the data and performed necessary cal-
`culations. When the calculations were complete,
`the
`computer printed out answers, and the operator took
`steps to set up the next problem.
`Second-generation computer. This series of oper-
`ations was inefficient, and the designers of second-
`generation equipment
`removed some of the ineffi-
`ciency by arranging input and output operations to be
`performed directly between the inpuVoutput periph-
`erals and the computer memory without
`interfering
`with computations, As a result, second-generation
`computers were able to perform computations while
`reading in data andprinting out replies, and efficiency
`was greatly enhanced. Figure 6 indicates schemati-
`cally the organizational change between generations
`of computers.
`:
`First-generation equipment was mostefficient while
`performing tedious and lengthy computations. The in-
`put/output capabilities of the second generation made
`
`IPR2018-01474
`Apple Inc. EX1022 Page 9
`
`|——_-+operator timei ef 1 preparation prepare SS)
`
`
`
`
`
`IPR2018-01474
`Apple Inc. EX1022 Page 9
`
`€
`

`

`290
`
`Digital computer
`control
`
`“computation
`a
`
`(b)
`
`Fig, 6. Evolution of organization for four generations of
`computers: (a) first; (b) second and third: (c) fourth,
`
`in applications where large volumes of
`them useful
`data had to be handled with relatively little computa-
`tion—applications such as billing. payroll, and inven-
`tory control. At the same time,
`the great capabilit
`and increased reliability of second-generation caer
`encouraged engineers to apply them to Situations
`where the computer acts as a control element.
`In mil
`itary aircrafl,
`in oil refineries and chemical plants in
`research laboratories, and in factories,
`the com itch
`received data directly from measuring instru Se
`performed appropriate calculations, and as aaeert
`made adjustments in the aircraft engine ina
`flow of raw materials in the plant or factory or te
`experimental setup in the laboratory. These
`é
`plication areasled to two important develo faa aan
`Computer design. The first was a new